Distributed Multi-Target Tracking for Autonomous Vehicle Fleets

Shorinwa, Ola; Yu, Javier; Halsted, Trevor; Koufos, Alex; Schwager, Mac

doi:10.48550/arxiv.2004.05965

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using the same consensus algorithm, [13] proposes a tracking approach in a distributed camera network. They address data association within each camera through a global metric, merging appearance and geometry cues, and acrossview data association through the euclidean distance between the 3D position of the targets The improvement of the Consensus Kalman Filter is discussed in [14] posed as a Maximum A Posteriori (MAP) optimization problem. The algorithm consists of closed-form algebraic iterations that guarantees the convergence to the centralized MAP over a designated sliding time window.…”

Section: B Multi-view Multi-target Tracking In Distributed Systemsmentioning

confidence: 99%

Distributed Multi-Target Tracking in Camera Networks

Casao¹,

García-Bosque²,

Murillo³

et al. 2020

Preprint

View full text Add to dashboard Cite

Most recent works on multi-target tracking with multiple cameras focus on centralized systems. In contrast, this paper presents a multi-target tracking approach implemented in a distributed camera network. The advantages of distributed systems lie in lighter communication management, greater robustness to failures and local decision making. On the other hand, data association and information fusion are more challenging than in a centralized setup, mostly due to the lack of global and complete information. The proposed algorithm boosts the benefits of the Distributed-Consensus Kalman Filter with the support of a re-identification network and a distributed tracker manager module to facilitate consistent information. These techniques complement each other and facilitate the cross-camera data association in a simple and effective manner. We evaluate the whole system with known public data sets under different conditions demonstrating the advantages of combining all the modules. In addition, we compare our algorithm to some existing centralized tracking methods, outperforming their behavior in terms of accuracy and bandwidth usage.

show abstract

Section: B Multi-view Multi-target Tracking In Distributed Systemsmentioning

confidence: 99%